Merge pull request #2395 from moj-analytical-services/add_symmetrical…

…_arg_to_columns_reversed

Fix ColumnsReversedLevel

CHANGELOG.md

@@ -11,6 +11,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 - Match weight and m and u probabilities charts now have improved tooltips ([#2392](https://github.com/moj-analytical-services/splink/pull/2392))
 
+### Fixed
+
+- Fixed issue where `ColumnsReversedLevel` required equality on both columns ([#2395](https://github.com/moj-analytical-services/splink/pull/2395))
+
 ## [4.0.1] - 2024-09-06
 
 ### Added

splink/internals/comparison_level_library.py

@@ -333,31 +333,42 @@ def __init__(
         self,
         col_name_1: Union[str, ColumnExpression],
         col_name_2: Union[str, ColumnExpression],
+        symmetrical: bool = False,
     ):
         """Represents a comparison level where the columns are reversed. For example,
         if surname is in the forename field and vice versa
 
+        By default, col_l = col_r.  If the symmetrical argument is True, then
+        col_l = col_r AND col_r = col_l.
+
         Args:
             col_name_1 (str): First column, e.g. forename
             col_name_2 (str): Second column, e.g. surname
+            symmetrical (bool): If True, equality is required in in both directions.
+                Default is False.
         """
         self.col_expression_1 = ColumnExpression.instantiate_if_str(col_name_1)
         self.col_expression_2 = ColumnExpression.instantiate_if_str(col_name_2)
+        self.symmetrical = symmetrical
 
     def create_sql(self, sql_dialect: SplinkDialect) -> str:
         self.col_expression_1.sql_dialect = sql_dialect
         self.col_expression_2.sql_dialect = sql_dialect
         col_1 = self.col_expression_1
         col_2 = self.col_expression_2
 
-        return (
-            f"{col_1.name_l} = {col_2.name_r} " f"AND {col_1.name_r} = {col_2.name_l}"
-        )
+        if self.symmetrical:
+            return (
+                f"{col_1.name_l} = {col_2.name_r} AND {col_1.name_r} = {col_2.name_l}"
+            )
+        else:
+            return f"{col_1.name_l} = {col_2.name_r}"
 
     def create_label_for_charts(self) -> str:
         col_1 = self.col_expression_1
         col_2 = self.col_expression_2
-        return f"Match on reversed cols: {col_1.label} and {col_2.label}"
+        direction = "both directions" if self.symmetrical else "one direction"
+        return f"Match on reversed cols: {col_1.label} and {col_2.label} ({direction})"
 
 
 class LevenshteinLevel(ComparisonLevelCreator):

splink/internals/comparison_library.py

@@ -1052,7 +1052,9 @@ def create_comparison_levels(self) -> List[ComparisonLevelCreator]:
             )
 
         levels.append(
-            cll.ColumnsReversedLevel(forename_col_expression, surname_col_expression)
+            cll.ColumnsReversedLevel(
+                forename_col_expression, surname_col_expression, symmetrical=True
+            )
         )
 
         for threshold in self.jaro_winkler_thresholds: